454 CHEMICAL ENGINEERING expected if heavy clouds of smoke are in the line of sight one day and not on the next day. If the pyrometer is sighted on a stream of molten iron during pouring or tapping the surface of the metal is usually clear and free from oxide. If the stream should at any time contain much slag, the 'surface will show bright patches on account of the higher emissivity of the slag. To make the readings conform with those taken on the clear stream, one must sight on the darker spaces between the slag or sight upon the slag and correct both sets of data according to Table 12. This table was computed from the following equation where # is the true absolute temperature, S the observed absolute temperature and E\ the emissivity for the wave length X. This wave length has been selected as X = 0.65/z, the approximate value for optical pyrometers. * _ i ^Mog-fl* = log E\ & S 6,232 9,588 The following table gives the emissivity of various materials for this wave length. The change of emissivity with temperature is usually small for metals. TABLE 13.—MONOCHROMATIC EMISSIVITY FOR RED LIGHT (X — ABOUT 0.65/x) Material E\ - o.65/< Material E\ = 0.65/x Silver 0.07 0.13 0.22 0.33 0.38 0.33 0.37 0.11 0.15 0.60 0.48 0.46 0.43 0.41 0.95 0.90 0.80 Cuprous oxide. . ...... 0.70 0.98 0.95 0.92 0.96 0.85 0.37 0.36 0.30 0.30 0.95 0.85 0.25 to 0.5 n . , / soli< Golds v I hqu Platinum < Palladium ~ f s< Copper jj. Tantalum Tungsten < Nichrome i . . f 800°C ....... id .............. Iron oxide \ 1,000°C ....... solid . ( 1,200°C ......... x-r. , , , f 800°C liquid ........ Nickel oxide |MO()OC f solid ........ Iron (solid and liquid) \ liquid ......... Dlid quid Nickel (solid and liquid) ]3100°C ...... Indium 2,600°C ....... Rhodium 1,000°C Graphite powder (estimated) . . Carbon 2,000°C 3,000°C . . 600°C ..... Porcelain (??) ................ 900°C 1,200°C Table 14 shows the corrections which must be added to the readings obtained with an optical pyrometer using light of wave length X = 0.65ju, for various emissivi-ties, in order to obtain the true temperatures. These data are especially useful when carefully plotted with observed temperatures as abscissae and corrections as ordinates. A family of curves is thus obtained corresponding to the different values of the emissivity.